Add comprehensive test coverage for Python wrangles (format, select, compare)

[Copilot]

Python wrangles (functions called directly in Python code) had minimal test coverage. Format module had only 5 tests, while select and compare modules had none.

Changes

Added 51 tests across three modules:

Format module (21 tests)

• concatenate - list joining with separator control
• coalesce - first non-empty value selection
• remove_duplicates - case-sensitive/insensitive deduplication
• tokenize - boundary and custom function splitting
• flatten_lists - recursive list flattening

Select module (18 tests)

• highest_confidence - max confidence selection
• confidence_threshold - threshold-based fallback
• list_element - indexing and slicing
• dict_element - key extraction and renaming

Compare module (12 tests)

• contrast - string intersection/difference
• overlap - fuzzy string matching with ratios

Example

# Format: concatenate with empty value handling
wrangles.format.concatenate([['a', '', 'b']], '-', skip_empty=True)
# Returns: ['a-b']

# Select: dictionary key extraction with defaults
wrangles.select.dict_element([{'name': 'Alice'}], 'age', default='Unknown')
# Returns: ['Unknown']

# Compare: find non-matching characters
wrangles.compare.overlap([['hello', 'hallo']], non_match_char='*')
# Returns: ['h*llo']

Test count: 55 → 106 (93% increase)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>More tests for python wrangles</issue_title>
<issue_description>It seems that we do not have many tests for python wrangles (wrangles that are meant to be uses in python, not python recipe wrangles). Specifically, format only has one, but I am sure that there are many others that are not being tested properly. </issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes More tests for python wrangles #842

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.